Vous êtes sur la page 1sur 12

A Quasi-Experimental Study Examining

the Safety Profile and Comfort Provided


by Two Different Blanket Temperatures
Lori T. Sutton, BSN, RN, CPAN, CCRN, Faye S. Baker, BSN, RN, CPAN,
Nan J. Faile, MS, RN, CCRP, Abbas Tavakoli, DrPH, MPH, ME
Surgical patients are often covered with warm blankets to alleviate the
discomfort of feeling too cold in the perioperative environment. The
goal of this study was to provide evidence to guide institutional blanket
warming policies by examining blanket thermal behavior and blanket
temperature preference of postoperative patients. The hypothesis was
that 155 F blankets are safe for patient use and provide a higher level
of thermal comfort to perioperative patients than do 110 F blankets.
A sample of 156 adult participants was randomized to the intervention
group (n 5 76), who received 155 F blankets, or the control group
(n 5 80), who received 110 F blankets. Participants were covered
neck to toe with the blankets in the postanesthesia care unit, and measurements were obtained for 10 minutes. An infrared thermometer was
used to measure skin and blanket temperatures, and a numeric scale
was used to measure thermal comfort. Blanket cooling rates were examined along with the effect of blanket temperatures on participants skin
temperature, oral temperature, and thermal comfort rating. Mean blanket temperatures for both groups were less than 93 F two minutes after
application. The intervention group showed higher skin temperatures
and thermal comfort throughout 10 minutes of data collection. The
results of this study support the hypothesis that it is safe to cover surgical patients with 155 F blankets. Results also indicate a correlation
between 155 F blankets and higher skin temperature and thermal
comfort.
Keywords: thermal comfort, blanket, PACU, temperature, research.
2012 by American Society of PeriAnesthesia Nurses

FEELING COMFORTABLY WARM is an important aspect of overall comfort for patients in health
Lori T. Sutton, BSN, RN, CPAN, CCRN, is a Staff Nurse, PostAnesthesia Care Unit, Lexington Medical Center; Faye S.
Baker, BSN, RN, CPAN, is a Clinical Mentor, Perioperative Division, Lexington Medical Center; Nan J. Faile, MS, RN, CCRP, is
a Research Nurse Coordinator, Lexington Medical Center, West
Columbia, SC; and Abbas Tavakoli, DrPH, MPH, ME, is a Statistician, University of South Carolina College of Nursing,
Columbia, SC.
Conflict of interest: None to report.
Address correspondence to Lori T. Sutton, Lexington Medical Center, Phase I Post-Anesthesia Care Unit, 2720 Sunset
Boulevard, West Columbia, SC 29169; e-mail address:
ltsutton@lexhealth.org.
2012 by American Society of PeriAnesthesia Nurses
1089-9472/$36.00
doi:10.1016/j.jopan.2012.01.011

Journal of PeriAnesthesia Nursing, Vol 27, No 3 (June), 2012: pp 181-192

care facilities. Warmed cotton blankets are used to


provide thermal comfort to patients in various hospital departments as well as in outpatient centers
and physician offices. In the perioperative area, patients often complain of feeling too cold before
and after surgical procedures secondary to cool
ambient room temperatures, decreased physical
activity, and thin hospital gowns. This perception
of cold may or may not be associated with hypothermia, which is defined as a core temperature
at or below 96.8 F (36 C).1 For patients who perceive cold when normothermic (core temperature
greater than 96.8 F), warmed blankets are a passive
warming measure that increase the feeling of
warmth by interfering with heat loss from the
skin to the environment.2

181

182

It must be emphasized that passive warming with


cotton blankets is appropriate only for normothermic patients who express subjectively feeling cold.
Warmed blankets do not increase the core body
temperature. Anesthetized patients can become
poikilothermic3 and be at high risk for development of hypothermia. For such patients, active
warming with forced air devices and warmed fluid
infusions are indicated to maintain or restore normothermia.1
Many patients describe the feeling of being cold as
the worst aspect of the hospital experience and
more unpleasant than pain. Frank et al4 describe research indicating thermal comfort is influenced
more by skin temperature than by core temperature. They hypothesize this to be an adaptive mechanism that triggers heat-conserving behaviors such
as adding layers of clothing. According to this theory, the drive to achieve thermal comfort may actually prevent heat loss and subsequent hypothermia.
In addition to being unpleasant, the perception of
feeling cold is physiologically stressful and can trigger autonomic effects that increase postoperative
complications. A patient who feels too cold can experience shivering, which increases heart rate and
blood pressure.5
Warming cabinets are designed specifically to
warm blankets used for patient care. Historically,
warming cabinet temperatures have been set arbitrarily. The warming cabinet temperature setting
became controversial in 2005 when the Emergency Care Research Institute (ECRI) recommended that the temperature setting for cabinets
used to warm patient blankets be limited to a maximum of 110 F. The ECRI recommendation describes reports of patients receiving burn injuries
when they were covered with blankets that were
too hot. However, in the article recommending
the 110 F limitation, the ECRI offered no specific
details of injuries related to warmed blankets.6
The authors of this article contacted ECRI to request further details regarding injuries caused by
warmed blankets. The ECRI representative responded that no other information was available.
ECRI identified the practice of storing patient care
liquids, such as those used for infusions and surgical
irrigations, in blanket cabinets as the primary thermal injury hazard prompting the 110 F recommen-

SUTTON ET AL

dation. Because the thermal properties of water


cause it to be more dangerous than blankets, ECRI
research focused on establishing 110 F as a safe
temperature for infusion and irrigation fluids. Results indicated a danger of patient burns from sustained skin contact with bottles of liquid heated to
greater than 110 F; therefore, the recommendation
of that temperature as the maximum temperature
for heating patient care fluids. ECRI recommended
the same maximum temperature for blankets because they are frequently stored in the same cabinet
with fluids.6 Compliance with the ECRI 110 F temperature limit for blanket warming cabinets has
been recommended by nursing organizations including the Association of PeriOperative Registered
Nurses (AORN) and the American Society of PeriAnesthesia Nurses (ASPAN).7,8
The safety of warming blankets to higher than
110 F is supported by evidence: the known physiology of burn injuries, the known physics of heat
transfer from cotton blankets to skin, and previous
nursing research demonstrating rapid blanket
cooling on removal from the warmer. This evidence will be discussed in the following paragraphs.
Heat causes burn injury by causing coagulation
and circulatory stasis in tissue.9 Actual research
demonstrating temperature and exposure time
necessary to burn human skin is scarce. Most publications related to thermal injury cite the seminal
thermal injury research published by Moritz and
Henriques10 in 1947. Moritz and Henriques circulated heated water through tubes placed in contact with live pigskin and measured the length of
time required to cause skin injury at various temperatures. They found that 48 C (118.4 F) was
the lowest temperature that could cause injury
more severe than simple redness. Edlich et al9
cite a temperature/length of exposure relationship
depicted in Figure 1, based on Moritzs and Henriques 1947 findings. Edlich et al9 found that
111.2 F is a critical temperature: if maintained
for longer than 6 hours, it will cause a cutaneous
burn. Note that a temperature of 129.2 F (54 C)
must be maintained for 1 minute to cause injury
and 123.8 F (51 C) must be maintained for 15 minutes to cause injury.
In order for a cutaneous burn to occur, heat energy
must be transferred from a warmer substance to

BLANKET TEMPERATURE COMPARISON

183

Figure 1. Relationship between temperature and duration of exposure in development of full-thickness burn injury.
Reprinted with permission.9

the skin. The amount of heat energy transferred by


one substance to another depends largely on the
important physical property known as specific
heat. Specific heat refers to the amount of heat energy required to raise the temperature of one gram
of substance by one degree Celsius. Water can gain
or lose a large amount of heat energy with very little
change in its temperature because of its high specific heat. Even if the initial temperatures of two
materials are identical, the stored heat energy available from water is much more likely to produce a severe injury because the specific heat of water (the
most common cause of scald burns) is the highest
of all the gases, metals, and solids so far tested,
with the exception of ammonia and ether.9

Blankets are not generally heated to a temperature


as high as 93 C (200 F) as described in Moons calculation.2 Blog entries report warming cabinet
temperatures between 140 F and 180 F.11 Kosson
et al12 published a study describing the measurement of cooling rates for blankets heated to
145 F in warming cabinets. Budjoso13 describes
researching blankets heated to 150 F.

Moon2, a biomedical engineer, applies the physics


of heat energy transfer to the fluid and blanket
question. The specific heat of water and skin is
very similar because skin is mostly water. When
a container of water-based fluid is placed in contact with skin, the cooler substance will absorb
heat from the warmer substance until the temperatures meet halfway between the initial temperaturesthe only heat energy lost being to the air
or carried away by circulatory flow. Cotton blankets, on the other hand, differ from skin because
blankets are mostly cellulose. A large amount of
air contained in the fabric weave causes cotton
blankets to have approximately the same specific
heat as air. Moon2 explains how to use theoretical

Kosson et al12 measured the rate of cooling for


145 F blankets at specific intervals after removing
the blankets from a warming cabinet. Average
blanket temperatures ranged from 91 F to 118 F,
86 F to 105 F, and 85 F to 105 F at 0, 30, and 60
seconds, respectively. Importantly, after 30 seconds, none of the blankets were hotter than
105 F, less than the critical temperature of
111.2 F indicated by Edlich et al9 in Figure 1.
Although limited by small sample size and scientific rigor, the Kosson et al12 findings indicate
that the blankets immediately lose large quantities
of heat energy when exposed to ambient air and
do not remain hot long enough to burn patients.
The authors recommended that other nurses

physics to determine how heat transfers between


blanket and skin. By his calculation, if a 93 C
(200 F) blanket is placed on 33 C (91.4 F) skin,
the skin temperature will only rise to 33.8 C
(92.8 F), well below the skin temperature necessary to incur a thermal injury.

184

conduct similar studies to further establish the safe


temperature setting for blanket warming cabinets.
Budjosos13 study compared 110 F blankets with
150 F blankets. The study examined rate of blanket cooling and perceived warmth experienced
by a sample of three volunteer female health care
workers. Blankets starting at 110 F decreased to
a mean temperature of 103 F at application, 90 F
by 1 minute, and 83 F by 5 minutes. The 150 F
blankets decreased to a mean temperature of
138 F at placement, 100 F by 1 minute, and 84 F
at 5 minutes. By the end of the data collection
time, all blankets approached room temperature.
The author also measured perceived warmth and
found that participants rated perceived warmth
higher with the higher temperature blankets.13
Limitations noted with this study include small
sample size and possible participant bias.
After institutions complied with lowering blanket
temperatures to the 110 F recommendation, ECRI
received numerous complaints that 110 F blankets
lacked necessary warmth to relieve chilling of hospitalized patients. In 2009, responding to numerous
requests to increase maximum allowable blanket
temperature, ECRI issued a new recommendation
that blankets could be warmed to 130 F. ECRI stated
that the higher blanket temperatures were safe
because of increased awareness of the need to store
liquids at a lower temperature separately from
blankets. This revised ECRI recommendation does
not describe any research conducted to support
the selection of 130 F as the safe or optimum
temperature.14

Methods
The purpose of this study was to further examine the safety profile for warmed cotton blankets
and determine a temperature that provides thermal comfort that is acceptable and safe for the
patient. Two warming cabinet temperature settings were compared to determine baseline blanket temperature effect on blanket temperature
change, skin temperature change, and patient
comfort level.
This study was conducted in a 26-bed postanesthesia care unit (PACU) in a 384-bed metropolitan medical center. The perioperative area
performs approximately 16,000 surgeries per

SUTTON ET AL

year. Two warming cabinets were used, one with


a 155 F thermostat setting and one with a 110 F
thermostat setting. The cabinets both have upper
and lower doors. The blankets used for all phases
of this study are commonly referred to as bath blankets; they are thin, 100% cotton blankets stored
folded to 16-ply in the warming cabinets.
A newly purchased Fluke 63 laser-guided, infrared, instant read thermometer (Fluke Corporation, Everett, WA) was used to measure all
surface temperatures sampled. To measure the
temperature of an object with the Fluke 63,
one only needs to aim and squeeze a trigger.
The thermometer emits a laser to light the objects surface at the measurement point. Designed specifically for noncontact temperature
measurement, it displays a digital temperature
readout measured from infrared energy radiated
by the objects surface. This type of thermometer
was recommended as the most accurate method
for surface temperature measurement by our biomedical technician and has been used in prior
studies.12,13 The Fluke 63 was calibrated when
manufactured, and its accuracy is certified in accordance with standards set by the National Institute of Standards and Technology. Additionally,
the laser causes the participant no discomfort
because it cannot be felt.
Oral temperatures were measured with the Welch
Allyn SureTemp Plus oral thermometer (Welch
Allyn Inc, San Diego, CA). The oral temperature
route was chosen because ASPAN cites strong
evidence that it is the peripheral route that best
approximates core temperature.1 Welch Allyn
oral thermometers are used routinely for patient
temperature monitoring in our PACU and are
calibrated yearly by the in-house biomedical
department.
Thermal comfort was rated on a numeric scale
(Figure 2) from 0 (intense cold) to 10 (intense
warmth). This thermal comfort scale was adapted
from a scale used to research thermal comfort
provided by heated and cooled automobile seats.
This type of scale is commonly used in the environmental comfort industry.15 The thermal comfort scale is also similar to the perceived warmth
scale described by Budjoso.13 The scale was
shown and explained to participants at the time
of consent.

BLANKET TEMPERATURE COMPARISON


0
Intense
cold

185

10
5
Not to hot
or cold

Intense
warmth

covered with two overlapping room temperature


blankets. The blankets were quickly lifted, and
warmed blanket surface temperature was measured immediately on completion of layering,
then each minute for a total of 5 minutes.

Figure 2. Thermal comfort scale.

Three phases of blanket pretesting were conducted to demonstrate blanket safety to Nursing
Administration and the Institutional Review Board
before enrolling patients in the study. The phase 1
pretest investigated the maximum temperature
blankets would attain in each warming cabinet.
This pretest began at 7 p.m. one evening. A principle researcher emptied both cabinets and then
filled each cabinet with room temperature blankets
until full. The door was taped shut to prevent opening by other staff. An assistant researcher was
trained by the principle researcher to obtain hourly
temperature measurement of blankets in each cabinet for 12 hours. The cabinets were kept closed as
much as possible, then quickly opened, and temperature was measured at the top, middle, and bottom of each stack in each cabinet.
Pretest phase 2 measured the temperature of each
blanket in each cabinet after allowing time for
maximum temperature to be attained. This was
done to determine the variability of blanket temperatures with regard to location within the cabinet. First, temperature indicated by the cabinets
digital readout was recorded. Then temperature
of each inner cabinet wall was measured, followed
by measurement of the temperature of the top and
bottom of each folded blanket in the cabinet.
Pretest phase 3 assessed rate of cooling for blankets removed from the 155 F cabinet. In this phase
of testing, researchers began by estimating room
temperature by measuring the temperature of an
interior wall surface next to the stretcher being
used for testing. Temperature of the bed surface
was then measured. Blanket temperature was assessed just prior to removing the blanket from
the warmer. Two blankets were removed, and
the still-folded blankets were carried approximately 18 feet to a stretcher. Two primary investigators worked together to quickly unfold the
blankets on the bed until they were double thickness lengthwise and overlapped in the manner in
which they are used to cover patients in this
PACU. The warm blankets were then completely

The patient testing phase of this study was approved by the Institutional Review Board. Informed consent was obtained from each study
participant. A convenience sample of 156 participants took part in the patient testing phase. Our
statistician calculated that a sample of 100 to 150
would provide sufficient power to minimize the
chance of making a type II error. The researchers
used a quasi-experimental design to compare the
two temperature settings: 155 F (intervention
group, n 5 76) and 110 F (control group,
n 5 80). Participants were randomly allocated to
one of the two groups using a computerized random number generator. A coin flip was used to determine that even-numbered participants would be
the control group and odd-numbered participants
would be the intervention group.
Exclusion criteria included oral temperature below
96.8 F or greater than 100.4 F. These temperatures
indicate hypo- or hyperthermia as defined by ASPAN
standards and would require intervention to restore
normothermia.1 Participants were excluded if an
oral temperature could not be measured for any
reason or if unable to communicate with the
researchers as required for the thermal comfort assessment. For example, non-English speaking participants or unresponsive participants were excluded.
Other participants also were excluded because of
scheduling conflicts.
Data were collected during the morning or early afternoon. The warming cabinets were stocked in
the evening before data collection days to assure
that blankets had reached maximum temperature
prior to data collection. During data collection,
the upper cabinet was used to obtain blankets
only for study participants. The upper cabinets
were taped shut to prevent other staff from opening the cabinet and allowing unnecessary heat loss
from the blankets.
Each participant received standard preoperative
and intraoperative care, and data collection began
when the participant arrived in the PACU. First,
oral temperature was measured and the absence

SUTTON ET AL

186

Intervention Group
Gender Distribution

Control Group Gender


Distribution

Gender male

24

Gender female

56

Control Group Age


Distribution

Gender male

30
46

Gender female

Intervention Group Age


Distribution
18-30

18-30
31-50

8 6
28
38

31-50

10 6

51-70

28

32

51-70

Greater than 71

Greater than 71

Intervention Group
Type of Surgery

Control Group Type of


Surgery
Head and
Neck (5)

Vascular
(1)

Head and
Neck (2)

Urological
(1)

Urological
(1)

Neurological (12)

Neurological (14)

General
(27)
General
(30)

Orthopedic (15)
Orthopedic (10)

Gynecological (19)

Gynecological (19)

Figure 3. Blanket study demographic distribution.

of other exclusion criteria was verified by the


researchers. Baseline skin temperature was measured at a spot on the forearm, and the participant
was asked to rate numeric thermal comfort. The
researcher then obtained two blankets from the appropriate warming cabinet. All staff and the participants except the researchers were blinded to the
starting temperature of the blanket. The blankets
were quickly unfolded and placed directly on the
participant in an overlapping fashion, covering
the participant from neck to toe. The warm blankets were then covered with two room temperature blankets. Immediately, all blankets were
quickly lifted to measure forearm temperature (application measurement) at approximately the same
location as baseline. Then the blanket temperature

was measured from where it had been in contact


with the forearm. The arm was then recovered
with the blankets. Skin and blanket temperature
measurements were repeated at the same location
every minute thereafter for 10 minutes. Thermal
comfort rating was reassessed at 5 and 10 minutes.
Oral temperature was repeated at the end of the
study. Data collection for the participant was
then complete.
Statistical analysis was done on human testing
data. Categorical variables were analyzed using frequency distributions; in addition, graphs (pie and
line) were used to illustrate the characteristics of
control group and intervention group participants.
Data analyzed over time included oral temperature,

BLANKET TEMPERATURE COMPARISON

187

skin temperature, blanket temperature, and comfort level. For continuous variables, statistics were
used to compute measures of central tendency
(mean and median) and measures of spread (standard deviation and range). Because so many data
were being gathered over a span of time, General
Linear Model (GLM) analyses in SAS and MIXED
procedures (mixed linear model that allows statistical inferences about data) for continuous outcomes (skin temperature, blanket temperature,
oral temperature, and comfort level) were used to
examine the effects of time, group, and time by
group interaction. Data were statistically analyzed
at four points during the study: at baseline, application time, 5 minutes, and 10 minutes.
Analysis of time effect refers to the mean of all participants at time 1 (baseline) compared with all
participants at time 2 (blanket application). Analysis of group effect refers to the mean of all control
group participants at times 1 and 2 compared with
the mean of all intervention group participants at
times 1 and 2. Analysis of time by group interaction
effect refers to comparing each group at each time
with each other group at each other time.

Results
Phase 1 pretesting demonstrated that the blankets
warmed at different rates, depending on their location in the warming cabinet. Baseline temperature
of the blankets was 71 F. Table 1 summarizes the
measurements from phase 1 pretesting.

When the temperature of each blanket was measured in pretest phase 2, it was discovered that
there was variability of blanket temperature depending on location within the cabinet. These
data are summarized in Table 2.
Phase 3 pretesting showed that mean temperature
of blankets before removal from the 155 F warmer
was 148.8 F. The researchers were amazed to find
that in the time it took to carry the blankets to the
stretcher and open them, the mean temperature
had already dropped to 119.4 F, a decrease of
29.4 F. Within 30 seconds after opening, the
mean temperature of the blanket dropped to
107.6 F and by 1 minute to 104.5 F. At completion
of phase 3, after 5 minutes, mean blanket temperatures were 96.8 F.
Of the 156 participants enrolled in the human testing portion of the study, 80 were randomized to the
control group and 76 were randomized to the intervention group. Figure 3 illustrates demographic
details of the groups. Level of significance was
P #.05 for all statistical analyses.
The mean oral temperature for the control group
was 97.84 F at baseline (time 1) and 97.85 F at
10 minutes (time 2). The mean oral temperature
for the intervention group was 97.91 F at baseline
and 97.86 F at 10 minutes. Table 3 contains GLM
results and illustrates that there were no significant
differences in oral temperature means at any time
during data collection.

Table 1. Pretest Phase 1: Time to Maximum Temperature and the


Maximum Temperature Achieved

Top
Middle
Bottom

110 F cabinet

155 F cabinet

Upper Cabinet

Upper Cabinet

Time to Maximum
Temperature

Maximum
Temperature Achieved

Time to Maximum
Temperature

Maximum
Temperature Achieved

6h
9h
9h

112 F
113 F
111.5 F

2h
6h
6h

156.5 F
145 F
149 F

Lower Cabinet
Top
Middle
Bottom

8h
9h
6h

Lower Cabinet
109.5 F
111.5 F
105 F

9h
11 h
6h

158.5 F
148.5 F
144 F

SUTTON ET AL

188

Table 2. Variability of Blanket Temperature


in Cabinet

Thermostat
reading ( F)
Maximum wall
temperature ( F)
Minimum blanket
temperature ( F)
Maximum blanket
temperature ( F)

10

110 F Cabinet

155 F Cabinet

Upper
Cabinet

Upper
Cabinet

109

159

111

147

104

148

109

157

9
8
9

8
8

6
4

1
0
Baseline

109

158

111

151

94

129

111

148

Mean numeric thermal comfort rating was similar


for both groups at the beginning of data collection.
Mean control group comfort level was 4.80 at baseline (time 1), 5.01 at 5 minutes (time 2), and 5.14 at
10 minutes (time 3). Mean intervention group thermal comfort was 4.87 at baseline (time 1), 5.55 at
5 minutes (time 2), and 5.57 at 10 minutes
(time 3). Comfort rating increased for both groups
throughout testing, but the intervention group
reported a slightly higher mean comfort level at
5 and 10 minutes than the control group.
Figure 4 graphs minimum and maximum thermal
comfort for both groups, and Figure 5 graphs
mean thermal comfort for both groups.

Lower Cabinet Lower Cabinet


Thermostat
reading ( F)
Maximum wall
temperature ( F)
Minimum blanket
temperature ( F)
Maximum blanket
temperature ( F)

10
Thermal Comfort Scale Score

Item Measured

12

5 minutes

10 minutes

Control Group Maximum

Intervention Group Maximum

Control Group Minimum

Intervention Group Minimum

Figure 4. Minimum and maximum thermal comfort. This figure is available in color online at www
.jopan.org.

for comfort level. Table 4 details results of this


analysis.
The human testing phase correlated with pretesting phase results for blanket temperature decrease.
Blankets from both groups lost a great deal of heat
instantly on removal from the cabinet. Mean control group blanket temperature decreased from
111.04 F at baseline (time 1) to 90.65 F at application (time 2) and then to 87.78 F at 10 minutes
(time 3). Mean intervention group blanket temperature decreased from 152.97 F at baseline (time 1)
to 100.43 F at application (time 2) and then to
88.66 F at 10 minutes (time 3). Figure 6 graphs

GLM analysis shows that there was not significant


time by group effect for comfort level. There was,
however, significant time effect and group effect
Table 3. Oral Temperature Statistical Analysis
Relationship
Analyzed

P Value of Oral Temperature

Time
Group
Time by group

.5833
.5501
.3566

Figure 5. Average thermal comfort. This figure is


available in color online at www.jopan.org.

BLANKET TEMPERATURE COMPARISON

Table 4. Thermal Comfort Statistical Analysis


Relationship
Analyzed

P Value of Thermal Comfort

Time
Group
Time by group

.0451
.0018
.1767

control group blanket temperature changes, and


Figure 7 graphs intervention group blanket temperature changes.
GLM analysis compared baseline with time 1 and
showed that there were significant effects when
comparing blanket temperature means for all
time, group, and time by group interactions.
Table 5 summarizes analysis of blanket temperature
effects.
Intervention group participants had greater mean
skin temperature increase than did control group
participants. Overall, the mean skin temperature
for the control group increased from 87.68 F
at baseline (time 1) to 89.18 F at application
(time 2). The average skin temperature for the intervention group increased from 86.97 F at baseline (time 1) to 89.97 F at application (time 2).
Figure 8 illustrates minimum and maximum skin
temperatures, and Figure 9 illustrates mean skin
temperatures for each group at each minute.
Note that by the end of data collection, temperatures for both groups are similar.

Degrees Fahrenheit

GLM analysis compared baseline with time 1 and


showed that there was significant time effect for
skin temperature. We cannot, however, conclude

125
120
115
110
105
100
95
90
85
80
75

189

that there is a different change over time for the


treated group versus the control group. Detailed
P value analysis of skin temperature means are
listed in Table 6.

Discussion
Conducting this study in a busy PACU was a great
challenge for the researchers. Scheduling was complicated because the researchers had to preoperatively identify patients who would arrive in the
PACU during the time allotted for data collection.
The researchers subsequently had to identify
when those patients would be present for the
preadmission interview so that informed consent
could be obtained at that time. Sometimes, consenting and data collection were occurring in different
departments at the same time, requiring great coordination of efforts among the researchers.
There were 213 individuals consented for the
study and data were collected from 156 participants. Reasons for exclusion included scheduling
conflicts (27), surgery cancellation (8), fast-track
to Phase II (8), hypothermia (6), withdrawal from
study (4), inability to communicate (3), and data
collection error (1).
Although participants were randomly assigned to
intervention or control groups, resulting demographic distribution was very similar for the groups.
Similarity of the groups increases the validity of
the results for our population. The variety of age,
gender, and types of surgeries increases the ability
to generalize the results to the general adult population.

Maximum
Average
Minimum
Critical Temperature

Figure 6. Control group blanket temperatures. This figure is available in color online at www.jopan.org.

SUTTON ET AL

190

165

Degrees Fahrenheit

155
145
135
125
Maximum

115

Average

105

Minimum

95

Critical Temperature

85
75

Figure 7. Intervention group blanket temperatures. This figure is available in color online at www.jopan.org.

Oral temperatures were only measured for purposes


of establishing absence of study exclusion criteria.
However, the lack of significant difference of the
means reinforces the knowledge that warm blankets
do not significantly affect core body temperature.
Comfort level time by group analysis did not yield
significant results, but both time and group examination of the data did yield significant results. Importantly, when overall intervention comfort level
mean was compared with overall control group
comfort level mean, the intervention group
showed a significantly higher mean thermal comfort level than control group.
Blanket temperature for both groups decreased
very rapidly as soon as the blankets were taken
out of the warmer. Maximum temperature of
any blanket immediately after application to the
patient was 126 F for the intervention group
(mean, 100.43 F), and 1 minute later the maxi-

mum temperature had decreased to 108 F. The


108 F measurement is well below the critical
temperature of 111.2 F that would need to be
maintained for over 6 hours to cause a burn,
as noted by Edlich et al.9 These findings support
the researchers hypothesis that the intervention
group blankets are safe for patient use and do
not remain hot long enough to cause burn injuries.
The skin temperature increased for both groups
over time; the skin temperature for both groups
was significantly higher at time 1 as compared
with any group at baseline. Based on information
from Frank et al,4 the rapid skin temperature increase may be responsible for the sensation of
warmth experienced by individuals when covered
with a warm blanket. Although not significant,
when compared with the control group at the
same time, the slightly higher skin temperature
of the intervention group may also contribute to

Table 5. Differences of Least Squares Means for Blanket Temperature


Effect
Time
Group
Time 3 group
Time 3 group
Time 3 group
Time 3 group
Time 3 group
Time 3 group

Time

Group Status

Baseline
Baseline
Baseline
Baseline
Baseline
Baseline
1

Time

Group Status

P Value

Control
Control
Experimental
Control
Experimental
Control
Control

, .0001
, .0001
, .0001
, .0001
, .0001
, .0001
, .0001
, .0001

1
Experimental
Experimental
Experimental
Experimental
Control
Control
Experimental

Baseline
1
1
1
1
1

Degrees Fahrenheit

BLANKET TEMPERATURE COMPARISON

191

100
98
96
94
92
90
88
86
84
82
80
78
76

Intervention Group Maximum

Intervention Group Minimum

Control Group Maximum

Control Group Minimum

Figure 8. Minimum and maximum skin temperatures. This figure is available in color online at www.jopan.org.

the slightly higher thermal comfort rating they


reported.

Conclusion
The rapid reduction in blanket temperatures supports the hypothesis that intervention group blankets do not pose a burn hazard to postoperative
patients. The researchers conclude from these
findings that blankets intended for patient use
can be safely warmed in a cabinet set to 155 F.

Statistical calculations support the conclusion that


sample size provided sufficient power to the study.
The similarity in demographic distribution indicates that these results are generalizable to adult
surgical patients.
Skin temperature increased for all participants
from baseline to time 1, then to the end of sampling. The lack of significant time by group differences between control and intervention
groups at a given time indicates that the

Degrees Fahrenheit

92.0
91.0
90.0
89.0
88.0
87.0
86.0

Intervention Group Average

Control Group Average

Figure 9. Average skin temperatures. This figure is available in color online at www.jopan.org.

SUTTON ET AL

192

Table 6. Differences of Least Squares Means for Skin Temperature


Effect
Time
Group
Time 3 group
Time 3 group
Time 3 group
Time 3 group
Time 3 group
Time 3 group

Time

Group Status

Time

Baseline
Baseline
Baseline
Baseline
Baseline
Baseline
1

Group Status

P Value

Control
Control
Experimental
Control
Experimental
Control
Control

, .0001
.9148
.1728
, .0001
, .0001
, .0001
, .0001
.0695

1
Experimental
Experimental
Experimental
Experimental
Control
Control
Experimental

blankets from both groups exert a similar effect


on the skin.
There was significant difference in mean thermal
comfort analysis when all intervention group results were compared with all control group results.
Overall, intervention group participants experienced a higher thermal comfort level throughout
data collection.

Baseline
1
1
1
1
1

Assisting individuals to achieve optimal comfort


is one of the most important aspects of nursing
practice. This study supports the conclusion
that 155 F blankets are safe and provide
a higher level of thermal comfort to postoperative patients. Nurses should have the option of
providing 155 F blankets when doing so is assessed to be an appropriate intervention for patients.

References
1. Hooper VD, Chard R, Clifford T, et al. ASPANs evidencebased clinical practice guideline for the promotion of perioperative normothermia. J Perianesth Nurs. 2009;24:271-287.
2. Moon JK. Warmed blankets are safe. Available at:
http://www.enthermics.com/learn/pdfs/skin%20warming.pdf.
Accessed July 17, 2009.
3. Johnson JS, Loushin MK. The effects of anesthetic agents
on cardiac function. In: Iaizzo PA, ed. The Handbook of Cardiac
Anatomy, Physiology, and Devices. New York, NY: Springer
Science and Business Media; 2009:239.
4. Frank SM, Raja SN, Bulcao CF, Goldstein DS. Relative contribution of core and cutaneous temperatures to thermal comfort and autonomic responses in humans. J Appl Physiol.
2002;86:1588-1593.
5. Kurz A. Thermal care in the perioperative period. Best
Pract Res Clin Anaesthesiol. 2008;22:39-62.
6. Emergency Care Research Institute. Hazard report. Limiting
temperature settings on blanket and solution warming cabinets
can prevent patient burns. Health Devices. 2005;34:168-171.
7. Petersen C. Clinical issues: Ignoring product label warnings. AORN Journal. 2006;83:944-949.
8. Clifford T. Frequently asked questions. Breathline. 2008;
28:12-13.

9. Edlich RF, Bailey TI, Bill TJ. Thermal burns. In: Marx JH,
Hockberger RS, Walls RM, et al, eds. Rosens Emergency Medicine: Concepts and Clinical Practice. 5th ed. St. Louis, MO:
Mosby; 2002:801-813.
10. Moritz A, Henriques F. Studies of thermal injury. II. The
relative importance of time and surface temperature in the causation of cutaneous burns. 1947. Available at: http://
www.ncbi.nlm.nih.gov/pmc/articles/PMC1934304/pdf/amjpathol
00502-0018.pdf. Accessed July 29, 2009.
11. MacArthur S. Macs safety space. Available at: http://blogs
.hcpro.com/hospitalsafety/?s5ahh1the1fact1and1fiction1
about1blanket&x534&y56. Accessed December 5, 2009.
12. Kosson C, Tveit C, Belew J. How warm are your blankets?
Evidence challenges advisory. Breathline. 2009;29:6-7.
13. Budjoso PJ. Blanket warming, comfort and safety. AORN
Journal. 2009;89:717-722.
14. Emergency Care Research Institute. Hazard report
update ECRI Institute revises its recommendation for temperature limits on blanket warmers. Health Devices. 2009;38:
230-231.
15. Zhang YF, Wyon DP, Fang L, Melikov AK. The influence of
heated or cooled seats on the acceptable ambient temperature
range. Ergonomics. 2007;50:586-600.